Jump to content
Objectivism Online Forum

Leaderboard

Popular Content

Showing content with the highest reputation on 11/08/18 in all areas

  1. No, it is not that a third factor explains why the two correlate. It is that a third factor, a "independent variable" with relatively low poisitve correlation can trump the real extremely high correlation of a different "independent variable". Let X1, X2, X3,...Xn be independent variables, and let Y be the dependent variable. Let's assume that there are no other independent variables. And let's assume we can vary a single Xi while holding all other variables constant With this, let's say we find a very low positive correlation between X1 and Y, but a very high positive correlation between Xn and Y Under that scenario, X1 might still be the most important factor if we want to change Y.
    1 point
  2. As far as stats, they will never tell you causation. What you're thinking sounds more like they may find significant correlations, but significant does not mean important. Significant does not mean causal. What you're talking about is that a third factor could explain why the numbers correlate. The numbers might have nothing to do with each other even if their correlation is big and significant.
    0 points
×
×
  • Create New...